Morphology, Behaviour and Molecular Evolution of Giant Mouse Lemurs (Mirza Spp.) Gray, 1870, with Description of a New Species

Home , Cheirogaleidae, Dwarf lemur, Kirindy Forest, Mouse lemur

P.M. Kappeler et al.: A New Mirza Species

MORPHOLOGY, BEHAVIOUR AND MOLECULAR EVOLUTION OF GIANT MOUSE LEMURS (MIRZA SPP.) GRAY, 1870, WITH DESCRIPTION OF A NEW SPECIES.

Kappeler PM, Rasoloarison RM, Razafimanantsoa L, Walter L and Roos C

Key Words: Mirza, morphometrics, biogeography, phylogeny, mitochondrial cytochrome b, Microcebus, new species

Abstract

Giant mouse lemurs (genus Mirza) are small nocturnal primates endemic to Madagascar, of which a single species (M. coquereli) is currently recognized. It is distributed along Madagascar’s west coast, with a gap of several hundred kilometres between two presumed subpopulations. Previous studies in the field and in captivity indicated substantial differences in several aspects of the biology of these two subpopulations. We therefore collected morphometric, genetic and behavioural data from populations representing the southern and northern end of their range to examine these differences in more detail. We obtained standard morphometric field measurements and DNA samples from a total of 74 adult males and females at Kirindy (central western Madagascar) and Ambato (northwestern Madagascar) and compared their social organisation. We also studied a total of 9 Mirza specimens housed at the Rijksmuseum van Natuurlijke Historie Leiden (The Netherlands). Our morphometric analyses revealed that the two Mirza populations differed significantly in 12 out of 13 measures, with the northern Mirza sporting smaller values in all traits except testes volume. Northern Mirza spent the day in nests with 2-8 (mean 4.1) individuals, whereas Mirza in the south virtually always slept alone. Moreover, reproduction in the northern population occurred several months earlier than in the south. We also sequenced the complete mitochondrial cytochrome b (cyt b) gene from several specimens and found that (1) the two populations differed by 3.33-3.51 %, which is similar to genetic distances observed among several closely related species of mouse lemurs (Microcebus), (2) DNA extracted from tissue on skulls collected in 1868/1870 yielded partial cyt b sequences that aligned perfectly with the northern and southern population samples, respectively, and (3) Microcebus from Andasibe clearly differed genetically from all other known mouse lemur species, indicating a separate species status for this population. Based on the combination of morphological, behavioural and genetic differences between Mirza from Kirindy and Ambato we conclude that they should be separated at the species level. Because M. coquereli was described based on a specimen from the southern population, we describe the northern Mirza as a species new to science.

Introduction

Madagascar is a global hot spot for biodiversity and conservation (GOODMAN and BENSTEAD, 2003; MITTERMEIER et al., 1998; MYERS et al., 2000; SECH-

Primate Report 71, July 2005 3 P.M. Kappeler et al.: A New Mirza Species

REST et al., 2002; YODER et al., 2005). Because of the number of endemic taxa and their phylogenetic history and position, the primates of Madagascar (Lemurifor- mes), in particular, represent one of the top global primate conservation priorities (GANZHORN et al., 1997a,b; JERNWALL and WRIGHT, 1998). Information about the taxonomic status, geographical distribution and abundance of individual taxa constitute the necessary scientific basis for the development of effective conservation action plans and their legal implementation. In the case of lemurs, these basic data are still far from complete because many new taxa or new localities for known taxa continue to be described, even in the new millennium (GROVES, 2000; LOUIS et al., 2005; MAYOR et al., 2004; RASOLOARISON et al., 2000; THALMANN and GEISSMANN, 2002) and their phylogenetic relationships are far from being resolved (FAUSSER et al., 2004; NIEVERGELT et al., 2002; PASTORINI et al., 2001a,b, 2002a,b, 2003; ROOS et al., 2004; YODER et al., 2000). Ongoing field studies, genetic analyses and museum work all indicate that the full taxonomic diversity of lemurs is still incompletely described and that conservation priorities need to be updated constantly. The mouse and dwarf lemurs (Cheirogaleidae) represent the largest lemur family. All members of the five currently recognized genera (Allocebus, Cheirogaleus, Microcebus, Mirza and Phaner) are relatively small (< 500 g) and nocturnal (MAR- TIN, 1972). Several field studies of cheirogaleids initiated in the 1990s used exten- sive trapping, detailed morphometrics and various genetic tools to address questions about their behavioural ecology. These new kinds of data also revealed the existence of new species (SCHMID and KAPPELER, 1994; ZIMMERMANN et al., 1998) and prompted systematic taxonomic work on this group, using a combination of osteo- logical, morphometrical and genetic data (GROVES, 2000; PASTORINI et al., 2001b; RASOLOARISON et al., 2000; RUMPLER et al., 1998; YODER et al., 1996, 2000, 2002). Despite difficulties arising from many synonyms, missing and damaged holotypes or those with vague collection localities and descriptions based on lecto- types or neotypes, the previously single recognized species of west coast mouse lemur (Microcebus murinus) could be differentiated into seven different species, three of which were new to science (RASOLOARISON et al., 2000). A similar taxonomic revision of the genus Cheirogaleus, albeit based on analyses of museum specimens alone, has indicated the existence of several species of dwarf lemurs along Madagas- car’s east coast, where traditionally only one species (C. major) had been recognized (GROVES, 2000). The overdue re-evaluations of the taxonomic status and phylogenetic position of Phaner and Allocebus still await further field data (GROVES and TATTERSALL, 1991). The genus Mirza GRAY, 1870 is currently represented by a single recognized species: Coquerel's dwarf lemur, M. coquereli (GRANDIDIER, 1867). While mainly considered as a mouse lemur (e.g. PETTER and PETTER-ROUSSEAUX, 1979), it was resurrected as separate genus by TATTERSALL (1982) because of its larger size and various behavioural and morphological differences with Microcebus. The genus Mir- za is restricted to western lowland forests, where it appears to have a disjunct distribution, with a gap between subpopulations spanning several hundred kilometres (Fig. 1). However, neither the exact limits of remaining population centres, nor the actual and historical presence or absence of Mirza in the intermediate regions are known (MITTERMEIER et al., 1994; PETTER et al., 1977; TATTERSALL, 1982).

4 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

Today, Mirza is found in the Sambirano region, with the northern Mahavavy river as a possible northern boundary of this species’ range, whereas in the south, it is known to occur between the Parc National de Bemaraha and the Parc National de Zombitse-Vohi- basia, with a possible southern limit of their distribution somewhere north of the Onilahy river (KAPPE- LER, 2003). Most recently, Mirza has also been sighted in the Réserve Naturelle Intégrale Tsingy de Namo- roka (S. GOODMAN, pers. comm.), which is between these two distribution nu- clei. The behavioural ecology of Mirza has been studied both near the southern and northern ends of its geographical distribution. The first detailed description of the natural history of Mir- za was provided by PET- TER et al. (1971), who collected opportunistic observations near Beroboka, Fig. 1: Distribution of the genus Mirza (black: confir- north of Morondava. AN- med, grey: inferred). Dots indicate origin of analysed DRIANARIVO (1981) con- Mirza and Microcebus samples (see also Table 1). ducted the first systematic study of ranging and social behaviour in the northern population on the Ampasindava Peninsula near Am- pasikely, northwest of Ambanja. A study by PAGES (1978, 1980) in Marosalaza forest north of Morondava focused on social organisation and feeding ecology. Finally, an ongoing study in Kirindy Forest (CFPF) north of Morondava has provided data on the social and genetic structure of another southern population (KAPPELER, 1997a; KAPPELER et al., 2002). The largest captive population of Mirza, which is housed at Duke University Primate Centre, was the subject of several behavioural and physio- logical studies (STANGER, 1995; STANGER et al., 1995). This colony was estab- lished in 1982 with animals from the Ambanja region captured near Antamboro between the villages of Ampasimbary and Ampasindava (D. HARING, pers. comm.).

Primate Report 71, July 2005 5 P.M. Kappeler et al.: A New Mirza Species

These previous studies have indicated the possible existence of behavioural and morphological differences between the northern and southern populations, e.g. in sexual dimorphism, relative testes size and seasonality of reproduction (KAPPE- LER, 1997a,b, 2003; but see ALBRECHT et al., 1990), but a direct comparison has not been attempted. To investigate possible taxonomic differentiation that may un- derlie these differences, we initiated a comparative field study of the northern and southern population and examined museum specimens from both regions. We conclude that the observed behavioural, morphological and genetic differences merit separation of the two populations at the species level and describe the northern population of Mirza as a new species.

Materials and Methods

Fieldwork Field studies of Mirza were conducted in Kirindy Forest (20°40'S, 44°39'E), where the German Primate Centre maintains a permanent research station, and on the Ambato Peninsula, situated between the town of Ambanja and the island of Nosy Faly, where we found Mirza at the same site on Ermitage beach (13°25'S, 48°29'E) where ANDRIANARIVO (1981) conducted his study. At Kirindy, Mirza has been studied continuously since 1993, whereas we studied the Ambato population during three separate trips in March 1999, April 2000 and October 2000. The study area in the dry deciduous forest at Kirindy has been described in great detail elsewhere (KAPPELER, 1997a; SORG et al., 2003). At Ambato, Mirza was found in a highly de- graded 4 ha patch of forest at the tip of a small peninsula next to the Hotel Ermitage Plage. This small piece of forest, vegetationally referable to the Sambirano Domain (HUMBERT, 1955) also contained several prominent mango trees and was accessi- ble through a small set of foot trails. At both field sites, we set Sherman and Tomahawk live traps and baited them with pieces of banana, mango or pineapple to capture Mirza. In addition, several individuals at Ambato were captured by hand from their daytime nests by local assis- tants. Captured animals were briefly anesthetised with 0.1 ml "Ketanest 100" and subjected to standard morphometric measurements, including body mass, head length and width, head-body length, tail length and hind foot length (following RASOLOARISON et al., 2000; SCHMID and KAPPELER, 1994). Not all measurements were taken from all animals in order to avoid stressing of awakening individuals. In addition, a small piece of ear skin (2x2 mm) was removed and stored in 70 % ethanol for later DNA-extraction and genetic analyses. Some animals at both sites were fitted with a small radio-tag (Biotrack, UK) to facilitate subsequent location and behavioural observations, including nesting habits.

Museum work The National Museum of Natural History (Rijksmuseum van Natuurlijke Histo- rie, RMNH), Leiden, The Netherlands, houses several Mirza specimens (RMNH 39375-39390), which were collected in the late 19th century at the "Bords du Mou- roundava" [=Morondava] and the "Baie de Pasandava" [=Ampasindava]. We mea- sured and compared several cranial landmarks on their skulls, using measurements

6 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species employed by RASOLOARISON et al. (2000), and examined mounted specimens ex- ternally. Small pieces of dried tissue could be obtained from the base of the skull of some specimens, and we successfully isolated and amplified DNA from these samples for comparison with samples obtained during our field studies.

Statistical analyses Quantitative analyses of morphometric data were limited to adult individuals, i.e. those weighing more than 250 grams (KAPPELER, 1997a). Cranial measurements obtained at the RMNH could not be compared statistically because the skull of one of three northern specimens was heavily damaged and one was a subadult. We used t-tests to compare mean measurements from males and females or from combined samples of individuals from Kirindy and Ambato, respectively. We tested for possible interactions between sex and origin by using a 2-way ANOVA. Because of the large number of tests based on this data set, alpha was set at 0.01 for all tests to guard against Type I errors.

Molecular genetics Genetic analyses were based on two different sample types. Ear clips from eight wild-caught Mirza, representing populations from Kirindy and Ambato, were collected during field surveys and stored in 70 % ethanol before further processing. Tis- sue material from six museum specimens of Mirza from Morondava and Pasandava was obtained from the RMNH. DNA from the tissue materials was extracted with the QIAamp DNA Mini Kit as recommended by the supplier and stored at -20° C. The complete mitochondrial cytochrome b (cyt b) gene was PCR amplified (SAIKI et al., 1988) with the oligonucleotide primers CYT-L: 5’-AAT GAT ATG AAA AAC CAT CGT TGT A-3’ and CYT-H: 5’-AAC TGC AGT CAT CTC CGG TTT ACA AGA C-3’. Standard, wax mediated hot-start PCRs were carried out for 40 cycles, each with a denaturation step at 94° C for 60 sec., annealing at 60° C for 60 sec. and extension at 72° C for 90 sec., followed by a final extension step for 5 min.. Because of the expected difficulties to amplify longer fragments from museum material, we amplified only a 192 bp long fragment of the cyt b gene (position 525 to 716 of the complete gene), using primers 5’-ACA CGA TTC TTT GCA TTC CAC-3’ and 5’-AGT AGA AGT AGG AGA AAG AGG-3’ with PCR conditions as described above with the exception that the extension time was reduced to 30 sec.. The results of the PCR amplifications were checked by running an aliquot on a 1 % agarose gel, stained with ethidium bro- mide. Subsequently, PCR products were cleaned with the Qiagen PCR Purification Kit and sequenced on an ABI 3100-Avant sequencer using the BigDye Terminator Cycle Sequencing Kit from Applied Biosystems and the primers as indicated above. The respective sequences were deposited in GenBank and are available under the accession numbers DQ093169-DQ093182, DQ095782 and DQ095783. Sequences were easily aligned by eye due to the lack of insertions or deletions and checked for their potential to be correctly transcribed in order to eliminate data set contaminations with pseudogenes. For a comprehensive evaluation of the sequence data, we expanded our data set with self-generated sequences from two specimens of Microcebus rufus from Andasibe and with orthologous sequences already deposited at GenBank from most currently recognized Microcebus species (YODER et al., 2000), one individual of Mirza coquereli (YODER et al., 1996) and one individual of

Primate Report 71, July 2005 7 P.M. Kappeler et al.: A New Mirza Species

Allocebus trichotis (ROOS et al., 2004), which was used as outgroup for phylogenetic tree reconstructions. The final alignment comprised 24 sequences with 1140 bp in length. Further details about individuals and sequences are summarized in Fig. 1 and Table 1.

Table 1: Origin, sample type and GenBank accession number of analysed species for genetic studies.

species abbre- origin sample type GenBank viation Allocebus trichotis - - sequence AY441461 Microcebus murinus - Mandena sequence AF285565 M. murinus - Vohimena sequence AF285564 M. murinus - Kirindy sequence AF285561 M. murinus - Andranomena sequence AF285559 M. griseorufus - Beza Mahafaly sequence AF285568 M. ravelobensis - Ankarafantsika sequence AF285532 M. tavaratra - Ankarana sequence AF285534 M. berthae - Kirindy sequence AF285543 M. myoxinus - Aboalimena sequence AF285539 M. myoxinus - Bemaraha sequence AF285535 M. sambiranensis - Manongarivo sequence AF285556 M. rufus - Ranomafana sequence AF285551 M. simmonsi - Tampolo sequence AF285553 Microcebus sp. - Andasibe* feces DQ095782 Microcebus sp. - Andasibe* feces DQ095783 Mirza sp. - Antanboro, close to Pasandava sequence U53571 Mirza sp. Ambato 1 Ambato tissue DQ093169 Mirza sp. Ambato 2 Ambato tissue DQ093170 Mirza sp. Ambato 3 Ambato tissue DQ093171 Mirza sp. Ambato 4 Ambato tissue DQ093172 Mirza sp. Pasandava 10 Pasandava tissue,RMNH 39375 DQ093173 Mirza sp. Pasandava 11 Pasandava tissue,RNMH 39376 DQ093174 Mirza coquereli Kirindy 1 Kirindy tissue DQ093175 M. coquereli Kirindy 2 Kirindy tissue DQ093176 M. coquereli Kirindy 3 Kirindy tissue DQ093177 M. coquereli Kirindy 4 Kirindy tissue DQ093178 M. coquereli Morondava 12 Morondava tissue,RNMH 39385 DQ093179 M. coquereli Morondava 13 Morondava tissue,RNMH 39381 DQ093180 M. coquereli Morondava 14 Morondava tissue,RNMH 39380 DQ093181 M. coquereli Morondava 15 Morondava tissue,RNMH 39382 DQ093182 * individual kept at the Zürich Zoo, Zürich, Switzerland

8 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

Because of the upcoming description of three new Microcebus species from the east coast of Madagascar (LOUIS et al., 2005), a BLAST search was performed to identify the species affinity of the "M. rufus" individuals in our data set. Absolute pairwise differences within and between species and genera were calculated with PAUP 4.0b10 (SWOFFORD, 1999) and DnaSP 3.52 (ROZAS and ROZAS, 1998). Phylogenetic tree reconstructions based on complete cyt b sequences were carried out with the maximum-parsimony (MP), neighbor-joining (NJ) and maximum-likelihood (ML) algorithms as implemented in PAUP or TREEPUZZLE 5.0 (STRIMMER and VON HAESELER, 1996). For MP analyses, all characters were treated as unor- dered and equally weighted throughout. NJ and ML trees were constructed with the TrN + I + G model of sequence evolution, because it was selected as best fitting model with MODELTEST 3.06 (POSADA and CRANDALL, 1998). Relative support of internal nodes was performed by bootstrap analyses (MP and NJ) with 1,000 replica- tions or by the quartet puzzling support values on the basis of 1,000 puzzling steps (ML). To examine the existence of significantly different lineage-specific evolutionary rates observable in the data set, we performed a relative rate test with the RRTree program (ROBINSON et al., 1998) for all possible pairwise comparisons and using the Allocebus trichotis sequence as outgroup. Divergence dates were estimated with the r8s program, version 1.7 (SANDER- SON, 2003) on the basis of estimated branch lengths as deduced from the NJ reconstruction. Age calculation was conducted with the nonparametric method and Powell’s optimisation, with all other settings set by default. As calibration points we used the proposed 24.2 million years ago (mya) for the divergence between Mirza and Microcebus (YODER and YANG, 2004).

Results

Morphology Two sets of morphometric data were available for the present analyses: external measurements form individuals captured at Kirindy (26 adult females and 30 adult males) and Ambato (8 adult females and 10 adult males), as well as cranial measurements from museum specimens at RMNH (7 from Morondava and 2 from Pasan- dava). Descriptive statistics for measurements taken at Kirindy and Ambato are summarized in Table 2. Mirza from Kirindy were on average heavier than animals from Ambato (Fig. 2; 2-way ANOVA: Origin F1;70 = 4.38, p = 0.04; Sex F1;70 = 0.23, NS; Origin*Sex F1;70 = 4.56, p = 0.03); despite the fact that several Ambato females were pregnant. Because there was no significant sex effect on body mass, only results of comparisons of combined-sex samples are presented below. The two populations did not differ significantly in body length (t = 1.58, df = 24, p = 0.12), but in tail length (t = 11.0, df = 48, p < 0.001), head length (t = 3.29, df = 51, p < 0.001), head width (t = 5.97, df = 51, p < 0.001), canine height (t = 10.3, df = 55, p < 0.001), ear length (t = 11.0, df = 40, p < 0.001), hind foot length (t = 5.42, df = 49, p < 0.001), as well as the length of their femur (t = 2.07, df = 35, p < 0.05), humerus (t = 3.03, df = 36, p < 0.001), radius (t = 6.62, df = 35, p < 0.001) and tibia (t = 5.21, df = 35, p < 0.001). In all cases, Mirza

Primate Report 71, July 2005 9 P.M. Kappeler et al.: A New Mirza Species from Ambato had the smaller means (Table 2). Northern Mirza only had marginally larger testes (t = 1.91, df = 38, p = 0.06).

380 Fig. 2: Box plots depicting va- 360 riation in body mass among male and female Mirza from 340 Ambato and Kirindy.

320

300 Body mass (g) 280

260

240 Female Male Female Male AMBATO KIRINDY

Proportional differences between members of the two populations were most pronounced with respect to canine height, tail length, ear length and testes volume (Fig. 3). When we used head length to control for size effects, most variables scaled allometrically with a clear separation between northern and southern populations with little to no overlap (Fig. 4). The northern Mirza clearly is a scaled-down version of the southern Mirza.

-30 -20 -10 0 10 20 30 40 Fig. 3: Proportional differ- 1 ences between northern Body mass and southern Mirza in Body length 2 morphometric variables. 3 The northern population Tail length (Ambato) served as a refer- Head length 4 ence for the Kirindy population. Head width 5 Canine height6 Ear length 7 Hind foot 8 Femur 9

Tibia 10 Humerus 11 Radus 12 13 Testes volume

10 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

35 38

34 36

33 34

32 32

31 30 Ear length (mm)

Head width (mm) 30 28

29 26

28 24 50 51 52 53 54 55 56 57 58 59 60 61 52 53 54 55 56 57 58 59 60 61 Head length (mm) Head length (mm) Ambato Kirindy

5,5 36 5,25 5 34 ) 4,75

mm 32 ( 4,5 4,25 30 4

3,75 Tail length (cm) 28

anine height 3,5 C 3,25 26 3 2,75 24 50 51 52 53 54 55 56 57 58 59 60 61 50 51 52 53 54 55 56 57 58 59 60 61 Head length (mm) Head length (mm) Fig. 4: Allometry of the two Mirza populations. Head width and the three most devi- ant variables (ear length, tail length and canine height; see Fig 3) are scaled against head length for northern (grey) and southern (black) Mirza.

Behaviour Adult Mirza at Kirindy occupied home ranges of about 4 ha that overlapped to various extents with those of neighbours (KAPPELER, 1997a). They spent the day in self-constructed nests, most of which were built 1-2 m below the top of Securinega (Family Euphorbiaceae) trees (SARIKAYA and KAPPELER, 1997). Radio-collared adult males and females were never found to share a nest during the day (KAPPE- LER, 1997a). At Ambato, in stark contrast, we always found Mirza sleeping in groups of 2-8 (mean 4.1; N = 18 nests) individuals. Because many of these nests contained radio-collared or marked individuals, who were observed at dawn while leav- ing a nest, or because nests were emptied by hand during the day, we could identify the age/sex class of most individuals. We found that nests at Ambato contained on average 0.77 adult females, 1.06 adult males, 0.44 juveniles and 1.89 unidentified individuals. This form of gregariousness had already been noted by ANDRIANA- RIVO (1981) and is not due to a lack of nests in this disturbed habitat because i) we also captured 5 animals from a single nest in a much larger forest at the foothills of Ambato Massif several kilometres to the east, and ii) each radio-tagged individual used between 2-5 different nests on the 3-7 days they could be located. The social or- ganization of the Mirza populations at Kirindy and Ambato are therefore fundamen- tally different in this respect.

Primate Report 71, July 2005 11 P.M. Kappeler et al.: A New Mirza Species

Reproductive activity in the Morondava region is restricted to a few weeks in No- vember (PAGES, 1978, 1980; KAPPELER, 1997a). At Ambato, we captured 4 pregnant females with clearly palpable foetuses (3 singletons; 1 pair of twins) during late September and early October in 2000. Mating at Ambato therefore presumably takes place in July and August. Members of the captive population at the Duke Uni- versity Primate Centre reproduced year-round (STANGER et al., 1995).

Molecular genetics Complete mitochondrial cytochrome b gene sequences were generated from eight wild caught giant lemurs and two Microcebus individuals from Andasibe, as well as a 142 bp long fragment from six museum specimens representing the two Mirza populations. The short fragment displayed five diagnostic mutations to distinguish between the populations from Morondava-Kirindy and Pasandava-Ambato (Fig. 5). To obtain a comprehensive overview of Microcebus and Mirza evolution, we expanded our data set with orthologous sequences from all currently recognized Microcebus spp. and one Mirza individual, which were already deposited at Gen Bank (YODER et al., 1996, 2000). Mouse lemurs distributed along the east coast, traditionally comprised within M. rufus, were available from Ranomafana, Anda-

71 U53571 TACCTTTTATCATCACAGCCCTAGTAATAGTTCACCTCCTTTTCCTTCACGAAACAGGATCCAATAACCCA Ambato1 ...... Ambato2 ...... Ambato3 ...... Ambato4 ...... Pasandava10 ...... Pasandava11 ...... Kirindy1 ....C...... G...... G.C...... Kirindy2 ....C...... G...... G.C...... Kirindy3 ....C...... G...... G.C...... Kirindy4 ....C...... G...... G.C...... Morondava12 ....C...... G...... C...... Morondava13 ....C...... G...... G.C...... Morondava14 ....C...... G...... G.C...... Morondava15 ....C...... G...... G.C......

142 U53571 CTAGGTACCACATCAGACTCAGACAAAATCCCATTCCACCCATATTACACAATCAAAGACCTGCTAGGACT Ambato1 ...... Ambato2 ...... Ambato3 ...... Ambato4 ...... Pasandava10 ...... Pasandava11 ...... Kirindy1 .....C...... T...... Kirindy2 .....C...... T...... Kirindy3 .....C...... T...... Kirindy4 .....C...... T...... C...... Morondava12 .....C...... T...... Morondava13 .....C...... T...... C...... Morondava14 .....C...... T...... Morondava15 .....C...... T...... Fig. 5: 142 bp long alignment of the partial cyt b sequences including data obtained from museum material. Region represents position 554 to 695 of the complete cyt b gene with dots indicating identity with the reference sequence.

12 Primate Report 71, July 2005 Table 2: Descriptive statistics (mean ± SD) for 13 morphometric variables of adult male (M) and female (F) Mirza from Am- bato and Kirindy. Ambato Kirindy F M sexes combined F M sexes combined mean sd mean sd mean sd mean sd mean sd mean sd body mass 299.00 36.10 287.00 21.60 292.50 28.60 299.00 20.90 316.80 27.10 308.30 25.90 body length 24.10 0.96 24.50 0.41 24.28 0.75 24.90 1.92 25.03 1.11 25.01 1.27 tail length 27.90 1.30 27.10 1.41 27.27 1.38 31.30 1.08 31.81 1.29 31.69 1.23 head length 55.20 1.59 54.50 1.71 54.80 1.64 56.50 1.30 56.40 2.00 56.50 1.73 head width 30.90 1.25 30.10 0.97 30.40 1.15 32.10 0.91 32.50 1.15 32.30 1.07 canine height 3.32 0.28 3.63 0.12 3.50 0.25 4.47 0.37 4.60 0.38 4.55 0.38 ear length 27.10 1.72 28.30 0.91 27.80 1.41 32.70 1.32 33.60 1.72 33.10 1.63 hind foot 54.10 2.09 52.80 1.37 53.30 1.78 56.10 1.73 56.90 2.40 56.70 2.22 femur 58.10 2.45 59.60 1.16 58.70 2.03 60.40 3.33 61.70 3.22 61.20 3.32 tibia 67.20 3.14 69.60 1.71 68.20 2.74 73.10 2.66 74.90 3.06 74.12 2.98 humerus 44.40 2.67 44.70 0.90 44.50 1.98 46.20 2.44 47.40 1.82 46.90 2.12 radius 46.90 2.05 47.30 1.52 47.10 1.73 50.40 1.30 51.60 1.61 51.20 1.58 testes 21.60 5.29 17.65 5.79

Table 3: Minimum and maximum pairwise genetic distances (in %) within and among analysed species. 123456789101112 1 Mirza sp. (Ambato) 0.18-0.79 2 M. coquereli (Kirindy) 3.33-3.51 0.44-0.53 3 Microcebus berthae 16.40-16.67 17.11-17.37 - 4 M. myoxinus 16.67-17.54 16.75-17.37 4.30-4.65 1.32 5 M. ravelobensis 17.54-17.81 17.46-17.72 10.18 10.26-10.61 - 6 M. tavaratra 16.14-16.40 17.28-17.54 7.54 8.16-8.51 10.70 - 7 M. sambiranensis 16.40-16.58 16.93-17.19 6.75 7.72-7.90 11.29 8.33 - 8 M. simmonsi 16.32-16.58 17.02-17.28 6.49 7.81-7.98 11.49 8.33 7.02 - 9 M. rufus 15.97-16.14 16.29-16.49 3.86 4.30-4.83 9.91 7.72 7.28 6.82 - 10 M. sp. (Andasibe) 16.14-16.40 16.49-16.75 4.30 4.91-5.26 10.26 7.63 7.02 6.67 3.77 0.00 11 M. griseorufus 16.84-17.28 16.14-16.40 12.19 13.07 13.07 13.60 12.02 12.63 12.72 12.19 - 12 M. murinus 17.63- 17.19- 13.42- 14.12- 12.90- 14.47- 13.33- 13.16- 13.77- 12.98- 9.83- 0.61- 18.42 17.90 13.77 14.74 13.25 14.56 13.60 13.42 14.04 13.42 10.18 2.46 P.M. Kappeler et al.: A New Mirza Species sibe and Tampolo. The type location of M. rufus is "A few miles north of Fiana- rantsoa, central Betsileo", which is close to Ranomafana and hence, the individual analysed herein most likely represents M. rufus. BLAST search with sequences from the Tampolo individual revealed that this specimen corresponds to M. simmonsi (LOUIS et al., 2005) from Betampona and Zahamena. The individuals from Anda- sibe are closely related to specimens from the Parc National de Mantadia, but not referable to any known species and are therefore described as a new species below. We analysed individuals of M. berthae and M. murinus from different locations to compare their within-species variation with that observed between the Mirza populations. The average observed genetic differences among all sequences was 15.29 %, with the greatest differences being detected between genera (16.14-18.42 %). Among Microcebus spp., distances ranged from 3.77-14.74 %. The two Mirza populations differed in 3.33-3.51 %, which is similar to distances observed among several closely related Microcebus spp.. The average variations within the two Mirza populations from Ambato (n=4) and Kirindy (n=4) were 0.46 and 0.48 %, respectively (Table 3). Phylogenetic analyses were conducted with the maximum-parsimony, neighbor-joining and maximum-likelihood methods. All obtained trees showed the same topology and differed only in their support values for certain branches (Fig. 6). The monophyly of each of the two genera Mirza and Mircocebus was highly supported

Allocebus trichotis 100 Kirindy 1 100 100 Kirindy 2 100 Kirindy 3 100 Kirindy 4 Mirza 100 100 Ambato 1 100 Ambato 2 100 Ambato 3 Ambato 4 GenBank: U53571

Microcebus 75 100 M. myoxinus - Aboalimena 95 56 100 M. myoxinus - Bemaraha clade A 99 84 99 M. berthae 81 100 65 M. rufus 98 87 M. sp. - Andasibe 94 83 99 100 55 M. sambiranensis 95 86 M. simmonsi 100 M. tavaratra 100 M. ravelobensis 100 Microcebus M. murinus - Andranomena 100 100 96 81 98 M. murinus - Kirindy clade B 93 100 63 100 99 100 M. murinus - Vohimena 95 100 M. murinus - Mandena M. griseorufus Fig. 6: Phylogenetic relationships as obtained from complete mitochondrial cyt b gene sequence data. Numbers on nodes indicate support values for internal branches (top: MP, middle: NJ, bottom: ML). Microcebus is divided into two main clades (A and B).

14 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

(100 %). Within Mirza, the two populations from Kirindy and Ambato are clearly separated into two significantly supported clades. The genus Microcebus is also divided into two major groups with one comprising the larger bodied species, M. murinus and M. griseorufus (designated as clade B) and the other, with all the remaining, smaller bodied species (clade A). Within clade B, a major split occurred between M. griseorufus and M. murinus. In clade A, M. ravelobensis was the first to split off, followed by M. tavaratra. The remaining species were further separated into a clade consisting of M. sambiranensis and M. simmonsi, and a clade containing M. myoxinus, M. berthae, M. rufus and M. sp. nov.. Relationships within Microcebus were mainly resolved and statistically highly supported. Because significant rate differences were detected among lineages (data not shown), the calculation of splitting events was hampered by the absence of a molecular clock-like sequence evolution. To deal with these difficulties, time estimations were carried out using nonparametric methods that relax the stringency of the molecular clock assumption. Calculations were performed on the basis of branch lengths obtained from the NJ reconstruction in PAUP under the assumption of the TrN + I + G model of sequence evolution and applying the proposed 24.2 mya for the divergence between Mirza and Microcebus (YODER and YANG, 2004) as calibration point (Fig. 7). Accordingly, the initial split within the genus Microcebus occurred about 12.5 mya, which is in agreement with an earlier estimate of 12.0 mya (YODER

Allocebus trichotis U53571 Ambato 1 Ambato 2 Ambato 3 Mirza 2.1 Ambato 4 Kirindy 1 Kirindy 2 Kirindy 3 Kirindy 4 M. berthae 24.2 M. myoxinus - Aboalimena M. myoxinus - Bemaraha 2.2 M. rufus 5.7 M. sp. - Andasibe M. sambiranensis 8.8 M. simmonsi M. tavaratra Microcebus 12.5 M. ravelobensis clade A M. griseorufus 8.4 M. murinus - Mandena 1.4 M. murinus - Andranomena M. murinus - Kirindy clade B M. murinus - Vohimena Fig. 7: Estimation of the most recent common ancestors based on complete mitochondrial cyt b sequence data. The black circle indicates the divergence between Mirza and Microcebus 24.2 mya, which was used as calibration point. Grey circles and respective numbers refer to calculated divergence ages in mya.

Primate Report 71, July 2005 15 P.M. Kappeler et al.: A New Mirza Species and YANG, 2004). In clade A, M. ravelobensis split off from other species about 8.8 mya. Later on, two radiation-like separations among the remaining species occurred about 5.7 and 2.2 mya, respectively. In clade B, M. griseorufus diverged from M. murinus about 8.4 mya. The two Mirza populations from Kirindy and Ambato were separated about 2.1 mya which is comparable with the most recent splitting event between species of the Microcebus clade A (2.2 mya) and much earlier than the separation of M. murinus populations (1.4 mya).

Discussion and Conclusions

In this study, we analysed and compared the external morphology, behaviour, reproductive activity and genetic variability at a mitochondrial DNA locus of samples from the two presumed main populations of Mirza. Our analyses revealed several pronounced differences between these two populations. First, northern Mirza were significantly smaller in all but one external measurement (body length). Because our sample from Ambato included several pregnant females, future comparisons of non-pregnant females should therefore reveal even more pronounced differences in mean body mass. These differences in external morphology were most pronounced in ear length, tail length and canine size, which were all about 20-30 % shorter in animals from Ambato. Comparisons of all external measurements that controlled for differences in body size indicated northern Mirza to be scaled-down versions of the animals from the south. Pelage colour was not systematically recorded (cf. RASO- LOARISON et al., 2000), but variation within populations appeared to be as great as variation between populations. There may be a tendency for the pelage of northern animals to be less grey and slightly more reddish; their tails also appeared less dark towards the tip and their ventral parts were brighter. The mounted specimens at the RMNH were too faded to permit a meaningful direct comparison, however. The difference in ear size was the most pronounced and easily recognizable difference in external appearance between the two populations; also in mounted specimens. Second, as in the captive population at Duke University Primate Centre, northern Mirza males had larger testes than the animals at Kirindy (see KAPPELER, 1997a). Because our field measurements of testes size were taken months before or after the presumed mating season of northern Mirza, the present data presumably represent an underestimate of maximal testes size. As in virtually all other season- ally breeding lemurs, testes size of Mirza at Kirindy increased several-fold in the few weeks before the brief annual mating season (KAPPELER, 1997a). Because northern Mirza are smaller and because they have absolutely larger testes outside the mating season than Mirza at Kirindy, their maximal testes size may be the largest one in relation to body size among strepsirrhines (KAPPELER, 1997b). The differences in weaponry (canine size) and testes size between the two populations nicely correspond to each other and match theoretical expectations for primates with different levels of mate monopolisation (SETCHELL and KAPPELER, 2003). These data suggest that the northern Mirza is more promiscuous than the southern individuals, and they match the differences in nesting behaviour described above. Northern Mirza also reproduced months before the population at Kirindy. Be- cause of our limited sampling, we do not know whether reproduction in Ambato is

16 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species strictly seasonal, as in Kirindy, or aseasonal, as in captivity. Reproduction in lemurs is generally controlled photoperiodically and occurs earlier in taxa at higher lati- tudes (RASMUSSEN, 1985). Microcebus murinus, for example, which are found along much of the west coast reproduce several weeks earlier in Ampijoroa, compared to Kirindy about 500 km to the south (EBERLE and KAPPELER, 2004; RADESPIEL et al., 2002). Future studies therefore need to determine whether reproduction in northern Mirza is only advanced or aseasonal, compared to the southern population. Third, behavioural differences in daytime nesting were striking between northern and southern Mirza. Whereas the animals in Kirindy virtually always spent the day alone in their nests (apart from mothers and their offspring), we found on average more than 4 individuals at Ambato sharing a nest. We find it noteworthy that these nests tended to include several adult males with fully developed testes. In M. murinus, where the most detailed data on sleeping group composition exist, adult females form sleeping groups, whereas males typically sleep alone (RADESPIEL et al., 1999; WIMMER et al., 2002). It will therefore be interesting to study the social organisation of northern Mirza in more detail. Among other things, it will be interesting to determine whether co-sleeping females are also closely related (cf. KAPPE- LER et al., 2002). At Ambato, Mirza density based on census walks was calculated to be much higher (1086 individuals/km2) than at Kirindy (120 individuals/km2; KAPPELER, 1997a), and also higher than the 385 individuals/km2 estimated by ANDRIANA- RIVO in 1981. This high density may be due to the isolated status of this forest fragment, the presence of mango and other introduced food trees, and the fact that, except for a few groups of Eulemur macaco macaco, no other lemurs were present. To what extent Mirza competes with introduced Rattus, which was found in several Mirza nests, remains unknown. Thus, Mirza at Ambato occurs at relatively high density for unknown reasons, but so far they have only been studied at this one locality. Finally, variation in cyt b sequences within the two Mirza populations was of similar magnitude as variation within several Microcebus samples. Differences between the two Mirza populations were within the lower end of the range of pairwise differences determined for several other pairs of widely recognized Microcebus species. DNA obtained from the museum specimens at the RMNH fell squarely within the clusters of the respective field samples. We therefore conclude that the average genetic differences at this locus are comparable to differences observed between other pairs of closely related species. In addition, we found that the cyt b sequences of one Microcebus taxon differed with similar magnitude from those of its closest neighbours. We therefore conclude that these mouse lemurs need to be considered as a separate species. In conclusion, the two populations of Mirza compared in this study differ consis- tently and significantly in most external morphometric measures, they display radi- cally different social organisations and reproductive patterns, and they exhibit genetic differentiation indicative of typical species differences. We therefore conclude that the northern and southern populations of Mirza deserve to be separated at the species level. Because the type locality of Mirza coquereli (GRANDIDIER, 1867) is Morondava, the northern population needs to be named. We are unaware of any ex-

Primate Report 71, July 2005 17 P.M. Kappeler et al.: A New Mirza Species isting synonym for members of this genus (GROVES, 2001; HILL, 1953; SCHWARZ, 1931). We therefore describe the Mirza from northern Madagascar as a new species:

Mirza zaza sp. nov. KAPPELER and ROOS

Holotype: RMNH 39377, skull catalogued as "d" by JENTINK (1887) in his Cat- alogue ostéologique, skin as "c" by JENTINK (1892) in his Catalogue systématique of the Leiden mammal collections. An adult male, collected on 25th September 1865 at Congoni, Ampasindava Peninsula, by F.P.L. Pollen & D.C. van Dam. The skin is mounted. The skull is generally in good condition, except for a broken palate bone. No postcranial skeleton is available. Measurements (in mm; see definitions in RASOLOARISON et al., 2000): greatest skull length: 52.1; basal skull length: 43.3; greatest orbital diameter: 30.4; occipital width: 24.4; zygomatic breadth: 30.5; skull height: 20.4; maxillary canine height: 5.0; maxillary tooth row (P1-M3): 14.9. Paratypes: RMNH 39376, skull Cat. ost. "c", skin Cat. syst. "b", a subadult female, collected in 1868 in "Baie de Pasandava" (Ampasindava) by Pollen & van Dam. The skin is mounted. The skull is in good condition. RMNH 39375, skull Cat. ost. "b", skin Cat. syst. "a", an adult female, also collected in 1868 in "Baie de Pasandava" by Pollen & Van Dam. The skin is mounted, but the skull is greatly damaged. DNA from both specimens is stored at the Gene Bank of Primates, German Pri- mate Centre, Germany (GBP 1024 and 1025). Type locality: Madagascar: Province d’Ansiranana, "Baie de Pasandava" [= Am- pasindava], Congoni (13°40'S, 48°15'E) Description: Northern dwarf lemurs are covered with short greyish-brown fur that turns distinctly more grey ventrally. Hindlimbs are slightly longer than fore- limbs and locomotion is quadrupedal. The tail is long, bushy and darker towards the tip. Ears are relatively short and rounded. Diagnosis: Distinguished from Mirza coquereli by being generally smaller and by having relatively shorter ears, tails and canines (Fig. 8). Differs from M. coquereli in 3.33-3.51 % in the complete mitochondrial cytochrome b gene. Etymology: The name zaza is the Malagasy word for children. We chose this name for two reasons. First, it refers to the fact that the northern population is the more diminutive of the two species. Second, we wish to emphasize the responsibility of the current generation of Malagasy children for the conservation of this and other members of their fauna for future generations. Malagasy name: Tanta; English name: Northern giant mouse lemur; German name: Nördlicher Riesenmausmaki; French name: Microcèbe géaut du nord Distribution: As with many other newly described lemur species, the known range of M. zaza is essentially limited to the collection sites of Ambato and Pasan- dava. Intensive surveys are now required to obtain additional information about this species' distribution and abundance, so that potential study sites and conservation measures can be identified.

18 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

Fig. 8: Mirza zaza (top, Photo: D. Haring) and M. coquereli (bottom, Photo: M. Eber- le). Ear length is the most prominent external difference between the two species, with about 20 % shorter ears in M. zaza.

Primate Report 71, July 2005 19 P.M. Kappeler et al.: A New Mirza Species

Within the genus Microcebus, a large number of different species are recognized on the basis of morphological and molecular genetic data (LOUIS et al., 2005; PASTORINI et al., 2001b; RASOLOARISON et al., 2000, YODER et al., 2000). Mouse lemurs from Andasibe, however, are not referable to any known species. Since large genetic distances were detected between this population and other Microcebus spp. and no synonym is available for this population, we name this taxon as a new species:

Microcebus lehilahytsara sp. nov. ROOS and KAPPELER

Type Series: 9 alive specimens (6 males, 3 females) housed at the Zürich zoo, Switzerland. Specimens were caught at the type location by Samuel Furrer and Rob- ert Zingg in March 2005. DNA from all individuals is stored at the Gene Bank of Pri- mates, German Primate Centre, Germany (GBP 1033-1042). Type Locality: Madagascar: Province Toamasina, Andasibe (18°55’S, 48°25’E). Measurements: External head length: males (n=3): 33.5 (33.0-34.0) mm; females (n=1): 35.0 mm. Head-body length: males (n=3): 91.0 (90.0-92.0) mm; females (n=1): 90.0 mm. Body mass: males (n=6): 48 (38-64)g; females (n=3): 45 (30-54) g. Body mass data were collected in May 2005. Diagnosis and Description: M. lehilahytsara is one of the smaller-bodied mouse lemurs. Head-body length is similar to that in the smallest primate species M. berthae. The fur is dense and short, bright maroon with an orange tinge on the back, head and tail, turning creamy-white on the ventral side. A distinct white

Fig. 9: Male Microcebus lehilahytsara from the type locality Anda- sibe (Photo: R. Zingg).

20 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species stripe, extending from the upper end of the rhinarium to the lower forehead, is present. Ears are short and round. The tail is uniformly colored and used for storing fat. The scrotum is furred and testes are noticeably large (Fig. 9). The species differs from other Microcebus spp. in at least 3.77 % in the complete mitochondrial cytochrome b gene. Etymology: The species name "lehilahytsara" is a combination of the Malagasy words "lehilahy" and "tsara" which mean "man" and "good", respectively. We name this species in honour of Steven M. Goodman, who conducted numerous field surveys in Madagascar and provided important information about its biodiversity. By choos- ing this name, we express our appreciation for his crucial contributions to the description and preservation of Madagascar's lemurs and other animals. English name: Goodman's mouse lemur, German name: Goodman's Mausmaki; French name: MicrocPbe de Goodman Distribution: Currently, the species is only known from Andasibe and the Parc National de Mantadia. Further surveys are required to confirm the species' occur- rence in other areas.

Acknowledgements

We thank the Commission Tripartite of the MinistPre d’Environnement et Eaux et ForLts and Mme. Berthe Rakotosamimanana for their authorization and support of this study. We are much indebted to Steve Goodman for providing essential support, information and encouragement for the success of this study, as well as very constructive comments on this manuscript. We are indebted to Steve for his friend- ship, exemplary collegiality and hospitality. Special thanks go to Robert Zingg and the Zürich zoo for providing Microcebus samples and measurements, and for keep- ing the type series of M. lehilahytsara. We also thank Jenny Pastorini and the Parc Zoologique et Botanique de Tsimbazaza for further Microcebus samples from Anda- sibe. Thanks to Chris Smeenk at the National Museum of Natural History (Rijks- museum van Natuurlijke Historie), Leiden, for his cooperation and comments, and to Urs Thalmann for help in the identification of the museum specimens. We thank Dietmar Zinner, Claudia Fichtel, Ulrike Walbaum and Alexandra Dill for their help with the collection of morphometric and behavioural data during one field trip. Thanks also to David Haring, Manfred Eberle and Robert Zingg for contributing photos. This paper is dedicated to Mme. Berthe Rakotosamimanana on the occasion of her retirement from the Université d’Antananarivo to acknowledge her crucial role in the study and conservation of Madagascar’s fascinating lemurs over many de- cades, and to the memory of the late Jean-Jacques Petter, inspiring pioneer of mod- ern fieldwork on many lemurs, including Mirza.

References

Albrecht GH, Jenkins PD, Godfrey LR (1990): Ecogeographic size variation among the living and subfossil prosimians of Madagascar. Am J Primatol 22: 1-50.

Primate Report 71, July 2005 21 P.M. Kappeler et al.: A New Mirza Species

Andrianarivo AJ (1981): Etude comparée de l'organisation sociale chez Microcebus coquereli. Université de Madagascar, Antananarivo, unpublished Ph. D. thesis. Eberle M, Kappeler PM (2004): Sex in the dark: determinants and consequences of mixed male mating tactics in Microcebus murinus, a small solitary nocturnal primate. Behav Ecol Sociobiol 57: 77-90. Fausser JL, Andriaholinirina N, Rebarivola C, Rumpler Y (2004): Genetic comparison of two populations of Hapalemur simus inferred from D-Loop mitochondrial DNA sequences. Folia Primatol 75: 19-22. Ganzhorn JU, Langrand O, Wright P, O'Connor S, Rakotosamimanana B, Feistner A, Rumpler Y (1997a): The state of lemur conservation in Madagascar. Primate Cons 17: 70-86. Ganzhorn JU, Rakotosamimanana B, Hannah L, Hough J, Iyer L, Olivieri S, Rajaobelina S, Rodstrom S, Tilkin G (1997b): Priorities for biodiversity conservation in Madagascar. Primate Report 48-1: 1-25. Goodman SM, Benstead JP (eds, 2003): The Natural History of Madagascar. Chi- cago, University Press. Groves CP, Tattersall I (1991): Geographical variation in the fork-marked lemur Phaner furcifer (Primates, Cheirogaleidae). Folia Primatol 56: 39-49. Groves CP (2000): The genus Cheirogaleus: unrecognized biodiversity in dwarf lemurs. Int J Primatol 21: 943-962. Groves CP (2001): Primate Taxonomy. Washington, DC, Smithsonian Institution Press. Hill WCO (1953): Comparative anatomy and taxonomy. I. Strepsirhini. Edinburgh, University Press. Humbert H (1955): Les Territoires Phytogéographiques de Madagascar. Leur Carto- graphie. Colloque sur les Régions Ecologiques du Globe, Paris 1954. Ann Biol 31: 195-204. Jentink FA (1887): Catalogue ostéologique des mammifPres. Muséum d’Histoire Naturelle des Pays-Bas. Tome IX. E.J. Brill, Leide. Jentink FA (1892): Catalogue systématique des mammifPres. Muséum d’Histoire Naturelle des Pays-Bas. Tome XI. E.J. Brill, Leide. Jernvall J, Wright P (1998): Diversity components of impending primate extinc- tions. Proc Nat Acad Sci USA 95: 11279-11283. Kappeler PM (1997a): Intrasexual selection in Mirza coquereli: evidence for scram- ble competition polygyny in a solitary primate. Behav Ecol Sociobiol 41: 115- 128. Kappeler PM (1997b): Intrasexual selection and testis size in strepsirhine primates. Behav Ecol 8:10-19. Kappeler PM, Wimmer B, Zinner D, Tautz D (2002): The hidden matrilineal structure of a solitary lemur: implications for primate social evolution. Proc Roy Soc Lond B 269: 1755-1763. Kappeler PM (2003): Mirza coquereli, Coquerel's dwarf lemur. In: Goodman SM, Benstead JP (eds): The Natural History of Madagascar. Chicago, University Press: 1316-1318. Louis EE Jr, Coles MS, Andriantompohavana, R, Sommer JA, Mayor MI, Brenne- man RA (2005): Revision of the mouse lemurs (Primates, Microcebus) in east- ern Madagascar with a description of three new species.unpublished. Sequen- ces deposited at NCBI GenBank (http://www.ncbi.nlm.nih.gov).

22 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

Martin RD (1972): Adaptive radiation and behaviour of the Malagasy lemurs. Phil Trans Roy Soc Lond 264: 320-352. Mayor MI, Sommer JA, Houck M, Zaonarivelo J, Wright P, Ingram C, Engel S, Louis EE Jr (2004): Specific status of Propithecus spp. Int J Primat 25: 875-900. Mittermeier RA, Tattersall I, Konstant W, Meyers D, Mast R (1994): Lemurs of Mad- agascar. Washington, DC, Conservation International. Mittermeier RA, Myers N, Thomsen J (1998): Biodiversity hotspots and major tropi- cal wilderness areas: approaches to setting conservation priorities. Cons Biol 12: 516-520. Myers N, Mittermeier RA, Mittermeier C, da Fonseca G, Kent J (2000): Biodiversity hotspots for conservation priorities. Nature 403: 853-858. Nievergelt C, Pastorini J, Woodruff D (2002): Genetic variability and phylogeography in the wild Alaotran gentle lemur population. Evol Anthropol Suppl 1: 175-179. Pages E (1978): Home range, behaviour and tactile communication in a nocturnal Malagasy lemur Microcebus coquereli. In: Chivers DA, Joysey KA (eds): Recent Advances in Primatology, Vol 3. London, Academic Press: 171-177. Pages E (1980): Ethoecology of Microcebus coquereli during the dry season. In: Charles-Dominique P, Cooper HM, Hladik A, Hladik CM, Pages E, Pariente GF, Petter-Rousseaux A, Petter J-J, Schilling A (eds): Nocturnal Malagasy Pri- mates: Behavior, Ecology, and Physiology. New York, Academic Press: 97-116. Pastorini J, Forstner MRJ, Martin RD (2001a): Phylogenetic history of sifakas (Propithecus: Lemuriformes) derived from mtDNA sequences. Am J Primatol 53: 1-17. Pastorini J, Martin RD, Ehresmann P, Zimmermann E, Forstner MRJ (2001b): Mo- lecular Phylogeny of the lemur family Cheirogaleidae (Primates) based on mitochondrial DNA sequences. Mol Phylogenet Evol 19: 45-56. Pastorini J, Forstner MRJ, Martin RD (2002a): Phylogenetic relationships among Lemuridae (Primates): evidence from mtDNA. J Hum Evol 43: 463-478. Pastorini J, Forstner MRJ, Martin RD (2002b): Phylogenetic relationships of gentle lemurs (Hapalemur). Evol Anthropol Suppl 1: 150-154. Pastorini J, Thalmann U, Martin RD (2003): A molecular approach to comparative phylogeography of extant Malagasy lemurs. Proc Nat Acad Sci USA 100: 5879- 5884. Petter JJ, Schilling A, Pariente G (1971): Observations eco-éthologiques sur deux lémuriens malgaches nocturnes: Phaner furcifer et Microcebus coquereli. Terre Vie 25: 287-327. Petter JJ, Albignac R, Rumpler Y (1977): Faune de Madagascar 44: MammifPres Lémuriens (Primates Prosimien). Paris, ORSTOM and CNRS. Petter JJ, Petter-Rousseaux A (1979): Classification of the prosimians. In: Doyle GA, Martin RD (eds): The Study of Prosimian Behavior. New York, Academic Press: 1-44. Posada D, Crandall KA (1998): Modeltest: testing the model of DNA substitutions. Bioinformatics 14: 817-818. Radespiel U, Cepok S, Zietemann V, Zimmermann E (1998): Sex-specific usage patterns of sleeping sites in grey mouse lemurs (Microcebus murinus) in North- western Madagascar. Am J Primatol 46: 77-84.

Primate Report 71, July 2005 23 P.M. Kappeler et al.: A New Mirza Species

Radespiel U, dal Secco V, Drögemüller C, Braune P, Labes E, Zimmermann E (2002): Sexual selection, multiple mating and paternity in grey mouse lemurs, Microcebus murinus. Anim Behav 63: 259-268. Rasmussen DT (1985): A comparative study of breeding seasonality and litter size in eleven taxa of captive lemurs (Lemur and Varecia). Int J Primatol 6: 501-517. Rasoloarison RM, Goodman SM, Ganzhorn JU (2000): A taxonomic revision of mouse lemurs (Microcebus) occurring in the western portions of Madagascar. Int J Primatol 21: 963-1019. Robinson M, Gautier C, Mouchiroud D (1998): Sensitivity of the relative-rate test to taxonomic sampling. Mol Biol Evol 15: 1091-1098. Roos C, Schmitz J, Zischler H (2004): Primate jumping genes elucidate strepsirrhine phylogeny. Proc Nat Acad Sci USA 101: 10650-10654. Rozas J, Rozas R (1998): DnaSP version 3: an integrated program for molecular population genetics and molecular evolution analysis. Bioinformatics 15: 174-175. Rumpler Y, Ganzhorn JU, Tomiuk J, Leipoldt M (1998): A cytogenetic study of Microcebus murinus. Folia Primatol 69: 307-311. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA (1988): Primer-directed enzymatic amplification of DNA with a thermo- stable DNA polymerase. Science 239: 487-491. Sanderson MJ (2003): r8s: inferring absolute rates of molecular evolution and divergence times in the absence of a molecular clock. Bioinformatics 19: 301-302. Sarikaya Z, Kappeler PM (1997): Nest building behavior of Coquerel's dwarf lemur (Mirza coquereli). Primate Report 47: 3-9. Schmid J, Kappeler PM (1994): Sympatric mouse lemurs (Microcebus ssp.) in western Madagascar. Folia Primatol 63: 162-170. Schwarz E (1931): A revision of the genera and species of Madagascar Lemuridae. Proc Zool Soc Lond 1931: 399-426. Sechrest W, Brooks T, da Fonseca G, Konstant W, Mittermeier RA, Purvis A, Rylands A, Gittleman J (2002): Hotspots and the conservation of evolutionary history. Proc Nat Acad Sci USA 99: 2067-2071. Setchell J, Kappeler PM (2003): Selection in relation to sex in primates. Adv Stud Behav 33: 87-173. Sorg JP, Ganzhorn JU, Kappeler PM (2003): Forestry and research in the Kirindy Forest / Centre de Formation Professionnelle ForestiPre. In: Goodman SM, Benstead JP (eds): The Natural History of Madagascar. Chiacago, University Press: 1512-1519. Stanger K (1995): Vocalizations of some cheirogaleid prosimians evaluated in a phylogenetic context. In: Alterman L, Doyle G, Izard M (eds): Creatures of the Dark. New York, Plenum Press: 353-376. Stanger K, Coffman B, Izard M (1995): Reproduction in Coquerel´s dwarf lemur (Mirza coquereli). Am J Primatol 36: 223-237. Strimmer K, von Haeseler A (1996): Quartet puzzling: A maximum likelihood method for reconstructing tree topologies. Mol Biol Evol 13: 964-969. Swofford DL (1999): PAUP*: Phylogenetic analysis using parsimony (* and other methods), version v4.0b2. Sinauer Associates, Sunderland, MA. Tattersall I (1982): The Primates of Madagascar. New York, Columbia University Press.

24 Primate Report 71, July 2005 P.M. Kappeler et al.: A New Mirza Species

Thalmann U, Geissmann T (2000): Distribution and geographic variation in the western wooly lemur (Avahi occidentalis) with description of a new species (A. unicolor). Int J Primatol 21: 915-941. Yoder AD, Cartmill M, Ruvolo M, Smith K, Vilgalys R (1996): Ancient single origin for Malagasy primates. Proc Nat Acad Sci USA 93: 5122-5126. Yoder AD, Rasoloarison RM, Goodman SM, Irwing JA, Atsalis S, Ravosa MJ, Ganz- horn JU (2000): Remarkable species diversity in Malagasy mouse lemurs (Pri- mates, Microcebus). Proc Nat Acad Sci USA 97: 11325-11330. Yoder AD, Burns M, Génin F (2002): Molecular evidence of reproductive isolation in sympatric sibling species of mouse lemurs. Int J Primatol 23: 1335-1343. Yoder AD, Yang Z (2004): Divergence dates for Malagasy lemurs estimated from multiple gene loci: geological and evolutionary context. Mol Ecol 13: 757-773. Yoder AD, Olson LE, Hanley C, Heckman KL, Rasoloarison RM, Russell AL, Ranivo J, Soarimalala V, Karanth KP, Raselimanana AP, Goodman SM (2005): A mul- tidimensional approach for detecting species patterns in Malagasy vertebrates. Proc Nat Acad Sci USA 102: 6587-6594. Zimmermann E, Cepok S, Rakotoarison N, Zietemann V, Radespiel U (1998): Sym- patric mouse lemurs in North-West Madagascar: a new rufous mouse lemur species (Microcebus ravelobensis). Folia Primatol 69: 106-114.

Appendix Mirza specimens housed at the RMNH, Leiden.

Collection RMNH Skin Skull Sex Collection place Collecter date 39375 a b female Pasandava Bay 1868 FPL Pollen & DC van Dam 39376 b c female Pasandava Bay 1868 FPL Pollen & DC van Dam 39377 c d male Congoni 25.09.1865 FPL Pollen & DC van Dam 39378 d e male Mouroundava River 1870 DC van Dam 39379 e f male Mouroundava River 1870 DC van Dam 39380 f g male Mouroundava River 1870 DC van Dam 39381 g h male Mouroundava River 1870 DC van Dam 39382 h i female Mouroundava River 1870 DC van Dam 39383 i j female Mouroundava River 1870 DC van Dam 39384 j k female Mouroundava River 1870 DC van Dam 39385 k l female Mouroundava River 1870 DC van Dam 39386 - a* female Mouroundava [River] [1870] DC van Dam 39387 l** - - - - DC van Dam 39388 m** - - - - DC van Dam 39390 n** - - - - DC van Dam * complete skeleton; ** preserved in alcohol

Primate Report 71, July 2005 25 P.M. Kappeler et al.: A New Mirza Species

Author´s address:

Peter M. Kappeler: Department of Behavioral Ecology & Sociobiology, German Primate Centre (DPZ), Kellnerweg 4, 37077 Göttingen and Department Socio- biology/Anthropology, University of Göttingen, Berliner Str. 28, 37073 Göt- tingen, Germany. Rodin M. Rasoloarison, Léonard Razafimanantsoa: Department of Behavioral Ecology & Sociobiology, German Primate Centre (DPZ), Kellnerweg 4, 37077 Göttingen and Université d'Antananarivo, Antananarivo, Madagascar. Lutz Walter: Primate Genetics, German Primate Centre, Kellnerweg 4, 37077 Göt- tingen, Germany. Christian Roos: Primate Genetics, German Primate Centre, Kellnerweg 4, 37077 Göttingen, Germany and Gene Bank of Primates.

Correspondence to: [email protected]

26 Primate Report 71, July 2005